Development of Wastewater Treatment Methods for Flotation Plants

Abstract

In this thesis, development of treatment methods for removing or decreasing amount of components in process water which negatively affect flotation performance has been studied. Two different sulfide ore samples provided by Polimetal Madencilik A.Ş. (Turkey) and Somincor Neves-Corvo mine (Portugal) were studied. In the first stage, flotation tests with water recirculation were performed in order to determine the amount of ions accumulating in process water. In the second stage, flotation tests were performed by using synthetic water samples prepared at different concentrations and the effect on performance was revealed. Then, water treatment methods have been investigated in order to eliminate the negative effect. The main treatment methods examined in this thesis are; chemical oxidation, activated carbon adsorption and ion exchange resins. In the final stage, flotation tests were performed by using treated water and its effect on flotation performance was determined.

According to the studies on Gediktepe Cu-Zn sulfide ore provided by Polimetal Madencilik A.Ş., it was observed that talc minerals in ore negatively affect flotation performance in Cu circuit. Pre-flotation stage was applied before Cu flotation in order to remove talc minerals. However, water recirculation tests revealed that residual collector ions in process water resulted in loss of copper in the pre-float stage. Therefore, Cu flotation performance in Cu flotation section was deteriorated. Activated carbon was used to remove the remaining collectors in water. Copper flotation was improved by activated carbon treatment. Activated carbon was regenerated by heating at high temperature.

The Cu-Pb-Zn sulfide ore sent from Somincor Neves-Corvo mine was investigated by similar methodology. Ca2+, SO42- and S2O32- were determined as major ions by water recirculation tests. Synthetic water samples were prepared in different concentrations by Taguchi experimental design methods. Thiosulfate and sulfate ions increased froth stability. Thiosulfate ions depressed sulfide minerals. The hydrogen peroxide oxidation method was used to remove thiosulfate ions. In this method, thiosulfate ions are oxidized to more stable sulfate ions. Hydrogen peroxide was added to the process water to reduce the concentration of thiosulfate. Flotation performance was restored by hydrogen peroxide treatment. On the other hand, the amount of sulfate increased with the oxidation of thiosulfate to sulfate. Although the sulfate increased froth stability, it did not adversely affect the flotation performance. However, gypsum (CaSO4) precipitates in the presence of sulfate and calcium ions at saturation levels. CaSO4 results in scaling problems in flotation circuits and as a result adversely affects the plant performance. Therefore, sulfate treatment is highly desirable in flotation plants.

In this thesis, ion exchange resins were used for sulfate treatment. Amberlyst A21, Amberlite IRA-67, Amberlite IRA-400 and Dowex 1X2 have been tested for the removal of sulfate ions. Optimum parameters such as pH, contact time and adsorption capacity of the resins were determined. In addition, regeneration methods of resins have been studied. In the light of these studies, it was found that Amberlyst A21 and Amberlite IRA-67 were effective in acidic waters. Amberlite IRA-400 and Dowex 1X2 have higher adsorption capacity and do not require pH adjustment. Resins were found to be regenerated with NaOH. The process water treatment was carried out at the Somincor Neves Corvo plant using the Amberlite IRA-400 and Dowex 1X2. Studies conducted in the plant and laboratory tests done with synthetic water were close to each other. The flotation of the Somincor Cu-Pb-Zn sulfide ore was performed using Amberlite IRA-400 treated water and the effect of the method on flotation performance was investigated. In the Cu-Pb section, Cu recovery decreased and Pb recovery increased after treatment. In the Zn section, the negative effect of treatment was not observed.